The extant neuropsychological and neuroanatomic data suggest that age-related changes in the functioning of the medical temporal lobe memory system. Specifically, memory for context or source, which is reduced in patients with frontal lobe lesions, is also poorer in older compared to younger adults, and appears to show greater reductions with aging than simple fact or recognition memory. The latter component of memory appears to depend upon the medial temporal lobe system. The proposed experiments are designed to assess the relative contributions or the medial temporal and frontal lobe systems to age-related changes in memory function. Event-related brain potentials (ERPs) and neuropsychological test assessments will be obtained from young, middle-aged, and elderly adults. Specifically, to assess age-related changes in these two broad neural systems, a two-pronged approach is proposed: 1) since the novelty P3 component of the ERP appears to depend upon intact dorsolateral prefrontal cortex, and appears to reflect aspects of working memory (also linked to dorsolateral prefrontal cortex), modified versions of the novelty oddball paradigm will be used with the novelty P3 serving as "probe" of frontal lobe function; novel stimuli will be used to cue strategic changes that are known to be difficult for patients with frontal lobe lesions and normally aging elderly adults; and 2) studies of the contextual aspects of memory or source in comparison with fact or recognition memory will be performed, in which ERP and behavioral measures of encoding (subsequent memory effect) and retrieval (old/new repetition effect) will be recorded. convergent measures of functional indices (standardized neuropsychological instruments, e.g., Wisconsin Card Sorting; Wechsler Memory Scale) will provide the opportunity for obtaining age-associated double dissociations between these memory functions, their ERP concomitants, and their putative underlying neural systems. ERPs will be recorded form 30 scalp sites to enable current source density analyses, in order to resolve better scalp distributions to determine whether they and, by implication, their intracranial generators, show age-related change with respect to the memory functions that will be assessed. The data will be relevant to age-related changes in frontal and medical temporal lobe functioning, age-related differences in source and recognition memory, and their physiological underpinnings.